Get Started
Home
Topics
Search
Library
Research questionHow can unsupervised remote sensing detect unknown, complex changes without labeled examples?Remote sensing change detection must recognize changes that may be rare, complex, or unlike predefined categories. Without labeled training data, synthetic changes can encode assumptions that fail to transfer to unfamiliar change types.
AI
Computer Vision
Image & Video Processing
Machine Learning
Latest papersRecent research connected to this question, newest first.Make Some Noise: Unsupervised Remote Sensing Change Detection Using Latent Space PerturbationsThe paper presents MaSoN, an end-to-end framework that synthesizes changes in latent feature space using feature statistics from the target data during training. It reports results across five benchmarks and discusses extension to SAR and multispectral data.research paper · Sep 1, 2026
Related questions
How can point-supervised change detection produce reliable pixel-level changes from sparse annotations?How can remote-sensing vision-language models align heterogeneous sensor imagery with text beyond RGB?How can we fuse low-resolution hyperspectral and high-resolution multispectral observations without losing spatial or spectral detail?How can semantic occupancy systems detect unknown objects under distribution shift in autonomous-driving scenes?